Scientists of the Silesian Medical University took part in the international project to implement artificial intelligence technologies in ophthalmology. The result of the study was the EyeFM model, which demonstrates a significant increase in the accuracy of diagnoses and the effectiveness of clinical decisions.
According to the press secretary of the university Agatha Pustulka, the results were published in the journal Nature Medicine. The system was trained on 14.5 million ophthalmic images and 400,000 clinical texts from international databases.
Verification of the model included a retrospective analysis, testing with the involvement of specialists from different regions, and also a randomized double-blind study with the participation of 668 patients and 16 doctors. The use of EyeFM made it possible to increase the accuracy of diagnosis from 75.4% to 92.2%, and the correctness of decisions about further treatment – from 80.5% to 92.2%.
According to Professor Adrian Smedovsky, the head of the research group, the system helps doctors make more standardized reports, and patients more often follow therapeutic recommendations. Another advantage of EyeFM was the use of more affordable visualization methods in cases where expensive procedures were previously required.
The publication in Nature Medicine, one of the world’s leading medical publications, emphasizes the importance of research and its potential for clinical practice.
